In recent years there have been numerous calls for making the operations of international organizations more “transparent.” One element in these demands involves the idea that international negotiations should be open to the same level of outside scrutiny that is presumed to prevail with bargaining in domestic contexts. While transparency of this sort may have clear benefits by facilitating attempts to hold officials accountable, scholars have made less effort to consider whether making international bargaining more public might also have detrimental effects. I develop a game-theoretic model that provides four hypotheses about the relative benefits of open-door versus closed-door bargaining, and about the preferences of different actors with regard to this type of transparency. This model, which can be applied to international and domestic contexts, helps extend positive theories about the design of institutions while also providing insights for the normative question of when transparency is desirable. I show that the hypotheses developed are supported both by historical evidence from eighteenth-century disputes about publicity in national parliaments, and by evidence from the more recent dispute about making European Council of Ministers deliberations public.I would like to thank Simon Hix, Bernard Manin, Lisa Martin, John Odell, Andrea Prat, Ken Scheve, Karen Smith, Andrew Walter, Peter Wilson, seminar participants at the LSE and at Sciences-Po in Paris, as well as two anonymous referees for helpful comments and suggestions.